Wall Street delivered a warning Thursday. Two days after Democrats swept the midterm elections, the Nasdaq plunged 5.52%, the S&P 500 fell 4.36%, and Nvidia lost more than 9%. Roughly $3 trillion in U.S. stock market value disappeared as investors questioned the enormous sums poured into artificial intelligence.

The election was not the only cause. Disappointing semiconductor earnings and doubts about future AI profits fed the selloff. But the Democratic victories added a new uncertainty: whether Washington and newly elected governors will impose restrictions that slow the construction of America’s AI infrastructure.

The stakes reach beyond corporate profits. America is competing with China for technological leadership, and AI now underpins prosperity, cybersecurity, and military preparedness. Can Washington govern this technology without crippling America’s ability to build it?

A New Political Obstacle

When the new Congress convenes in January, Democrats will control the committees that investigate technology companies and write AI legislation. Their priorities are already taking shape.

On October 28, Representatives Jamie Raskin (D-Md.), Frank Pallone (D-N.J.), and Zoe Lofgren (D-Calif.) asked five leading AI developers to notify Congress of serious safety incidents within 48 hours of discovery, or within 24 hours if critical infrastructure appears involved. The request is voluntary for now, but it signals the oversight Democrats intend to pursue.

That instinct has merit, because today’s AI systems carry risks private companies should not be permitted to dismiss. But Congress must separate safeguards against genuine dangers from requirements that only add cost and delay.

Workable models already exist. The bipartisan FRONTIER Act, introduced in July by Representatives Lori Trahan (D-Mass.) and Jay Obernolte (R-Calif.), would require independent audits, incident reporting, and continuing assessments, with obligations scaled to the developer’s size. Senator Maria Cantwell (D-Wash.) has outlined a framework built on enforceable standards and independent oversight. Both concentrate scrutiny where the greatest harm lies.

Oversight must also resist the temptation to settle political scores with individual executives. Congressional investigators must establish facts rather than presume guilt. Scripture commands, “Justice, and only justice, you shall follow” (Deuteronomy 16:20, ESV). That command binds the investigator as surely as the investigated.

The More Immediate Threat

While Washington debates how AI should operate, state and local governments are deciding whether its facilities can be built at all.

On November 3, voters across the country approved dozens of measures restricting data center development. Newly elected Democratic governors in Texas, Ohio, and Georgia campaigned on promises to pause construction or impose tougher conditions on developers.

According to Reuters, UBS analysts estimate that construction pauses could put roughly 30% of anticipated U.S. data center capacity additions in 2028 at risk. That is not a forecast of mass cancellations, but a substantial share of planned construction now faces uncertainty just as American developers race for computing capacity.

AI development requires enormous computing facilities, advanced semiconductors, dependable electricity, and private financing. Lost months cannot be recovered simply because demand remains strong.

Construction limits carry a second cost. Large technology companies already possess extensive computing infrastructure, while smaller competitors must build or lease new capacity. Restrictions would entrench the giants and shut out newcomers, an ironic result for policymakers who decry concentrated corporate power.

Yet communities have legitimate grievances. Families should not subsidize the electricity demands of wealthy technology companies, and developers should not expect automatic approval for projects that burden their neighbors. The bipartisan Ratepayer Protection Act passed the House 417-3 in September but has stalled in the Senate. It would have states consider standards that assign grid-expansion costs to the large customers responsible for them. That approach deserves a vote. Christian stewardship requires meeting the nation’s technological needs without shifting unfair costs onto ordinary families.

The Cybersecurity Dilemma

On November 5, OpenAI released GPT-6 Cyber to a limited group of approved customers and assigned the model its highest cybersecurity risk classification, “Critical.”

The same AI that helps authorized researchers find software vulnerabilities could also help malicious actors exploit them. OpenAI’s designation is its own assessment, not independent proof of catastrophic harm. Still, it shows why Washington needs targeted safeguards rather than a false choice between unrestricted deployment and blanket prohibition.

Good rules require testing, controlled access, and monitoring for the most dangerous capabilities without obstructing beneficial research.

China Is Not Waiting

A Reuters investigation published July 31 examined more than 80 Chinese academic papers and patents and found that researchers tied to China’s military and security institutions had used outputs from American AI models to train specialized Chinese systems. The technique, model distillation, transfers capabilities from powerful systems into smaller ones. Reuters identified applications in surveillance, military software, drone navigation, and cyber operations.

That finding exposes a strategic vulnerability. Even as America restricts China’s access to advanced chips, Chinese researchers can extract useful capabilities through American AI services. The United States must therefore sustain the infrastructure needed to lead in AI while keeping its most sensitive capabilities out of adversaries’ hands. Needless construction delays would weaken the first; lax safeguards would forfeit the second.

A Christian Responsibility to Govern Wisely

For Christians, the question is not whether technology companies should escape government oversight. They should not. Human beings are created in God’s image (Genesis 1:27), and their dignity outranks corporate convenience, political ambition, and technological achievement.

Government must protect citizens from genuine harm, ensure fair treatment, and defend the nation. That duty demands prudence. “The prudent sees danger and hides himself, but the simple go on and suffer for it” (Proverbs 22:3, ESV). Prudence means recognizing when well-intended restrictions create new dangers, including a weaker national defense.

Not every AI project serves the national interest, and market discipline should expose the wasteful ones. Government must judge proposed restrictions by whether they prevent genuine harm without imposing greater costs on communities, innovation, or national security.

The incoming Democratic majority should begin with three priorities: require meaningful safeguards for the most capable AI systems, protect communities from unfair infrastructure costs, and avoid restrictions that needlessly weaken American technological leadership.

America’s AI future depends on more than what its engineers invent. It depends on whether its elected leaders have the wisdom to govern what they build. China will exploit American hesitation wherever it can. Congress must protect the public without compromising the capabilities on which the nation’s prosperity and security depend.